AUTOMATED MESSAGE SENTIMENT ANALYSIS AND AGGREGATION

• US 20180089171A1
• Filed: 09/26/2016
• Published: 03/29/2018
• Est. Priority Date: 09/26/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Arquero, Gregg M., Dow, Eli M., Hossain, Syed F., Schaeffer, Joshua, Tang, Yunli
• Granted Patent: US 10,642,936 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/226 Delivery according to priorities
• CPC Class Codes: H04L 51/42 Mailbox-related aspects, e.g. synchronisation of mailboxes

Abstract

Techniques are provided for performing automated operations to analyze and prioritize incoming user messages. One or more indications of a plurality of messages sent to one or more recipient users is received. Based at least in part on one or more natural-language processing libraries, multiple messages of the plurality are analyzed. Analyzing the multiple messages includes determining one or more sentiments associated with each of the multiple messages, and generating, based at least in part on the determining of the one or more sentiments, aggregate sentiment information regarding one or more subsets of the multiple messages. Based at least in part on the analyzing of the received message, indications are provided of the aggregate sentiment information generated regarding one or more subsets of messages.

Claims

1. A computer-implemented method for providing aggregate sentiment information regarding user email messages, the method comprising:

receiving, by a hardware-implemented message sentiment classification and prioritization system comprising a hardware-implemented sentiment analysis engine and a hardware-implemented sentiment aggregation engine, one or more indications of a plurality of email messages sent to one or more recipient users, wherein the hardware-implemented message sentiment classification and prioritization system comprises a microprocessor of a special-purpose computer;

analyzing, by the message sentiment classification and prioritization system and based at least in part on one or more natural-language processing libraries, multiple email messages of the plurality of email messages, wherein analyzing the multiple email messages includes;

determining, by the message sentiment classification and prioritization system, one or more sentiments associated with each of the multiple email messages;

identifying one or more subsets of the multiple email messages, wherein each subset comprises a plurality of email messages and the email messages of each subset are complete email messages, the one or more subsets of the multiple email messages are based on a time period in which email messages were received and/or sent; and

generating, by the message sentiment classification and prioritization system and based at least in part on the determining of the one or more sentiments, aggregate sentiment information regarding the one or more subsets of the multiple email messages, wherein the aggregate sentiment information comprises an aggregation of sentiment information from the plurality of email messages of each subset; and

providing, by the message sentiment classification and prioritization system and for at least one subset of multiple email messages, one or more indications of the aggregate sentiment information generated regarding the at least one subset; and

displaying the one or more indications of the aggregate sentiment information via a display.

2. The computer-implemented method of claim 1 wherein the method is performed at least in part via a software application plug-in executed in conjunction with an email client associated with each of at least some of the one or more recipient users.

3. The computer-implemented method of claim 1 wherein generating the aggregate sentiment information regarding the subset of multiple email messages includes extracting one or more keywords associated with a sentiment that is determined to be associated with each of the email messages of the subset.

4. The computer-implemented method of claim 1 wherein providing the one or more indications of the aggregate sentiment information generated regarding the subset of multiple email messages includes displaying statistical data regarding one or more sentiments associated with the multiple email messages of the plurality.

5. The computer-implemented method of claim 1 wherein providing the one or more indications of the aggregate sentiment information generated regarding the subset of multiple email messages includes providing at least some of the aggregate sentiment information to one or more users other than the one or more recipient users.

6. The computer-implemented method of claim 1 wherein the method further comprises, subsequent to the generating of the aggregate sentiment information, receiving a search parameter and responding by providing an indication of one or more sentiments determined to be associated with the search parameter in accordance with at least some of the multiple email messages of the plurality.

7. The computer-implemented method of claim 1 wherein providing the one or more indications of the aggregate sentiment information generated regarding the subset of multiple email messages is performed periodically in accordance with configuration information associated with at least some of the one or more recipient users.

8. A non-transitory computer-readable medium having stored contents that, when executed, configure a computing system to perform a method for providing aggregate sentiment information regarding user email messages, the method comprising:

receiving, by a hardware-implemented message sentiment classification and prioritization system comprising a hardware-implemented sentiment analysis engine and a hardware-implemented sentiment aggregation engine, one or more indications of a plurality of email messages sent to one or more recipient users, wherein the hardware-implemented message sentiment classification and prioritization system comprises a microprocessor of a special-purpose computer;

analyzing, by the message sentiment classification and prioritization system and based at least in part on one or more natural-language processing libraries, multiple email messages of the plurality of email messages, wherein analyzing the multiple email messages includes;

determining, by the message sentiment classification and prioritization system, one or more sentiments associated with each of the multiple email messages;

identifying one or more subsets of the multiple email messages, wherein each subset comprises a plurality of email messages and the email messages of each subset are complete email messages, the one or more subsets of the multiple email messages are based on a time period in which email messages were received and/or sent; and

generating, by the message sentiment classification and prioritization system and based at least in part on the determining of the one or more sentiments, aggregate sentiment information regarding the one or more subsets of the multiple email messages, wherein the aggregate sentiment information comprises an aggregation of sentiment information from the plurality of email messages of each subset; and

providing, by the message sentiment classification and prioritization system and for at least one subset of multiple email messages, one or more indications of the aggregate sentiment information generated regarding the at least one subset; and

displaying the one or more indications of the aggregate sentiment information via a display.

9. The non-transitory computer-readable medium of claim 8 wherein the method is performed at least in part via a software application plug-in executed in conjunction with an email client associated with each of at least some of the one or more recipient users.

10. The non-transitory computer-readable medium of claim 8 wherein generating the aggregate sentiment information regarding the subset of multiple email messages includes extracting one or more keywords associated with a sentiment that is determined to be associated with each of the email messages of the subset.

11. The non-transitory computer-readable medium of claim 8 wherein providing the one or more indications of the aggregate sentiment information generated regarding the subset of multiple email messages includes displaying statistical data regarding one or more sentiments associated with the multiple email messages of the plurality.

12. The non-transitory computer-readable medium of claim 8 wherein providing the one or more indications of the aggregate sentiment information generated regarding the subset of multiple email messages includes providing at least some of the aggregate sentiment information to one or more users other than the one or more recipient users.

13. The non-transitory computer-readable medium of claim 8 wherein the method further comprises, subsequent to the generating of the aggregate sentiment information, receiving a search parameter and responding by providing an indication of one or more sentiments determined to be associated with the search parameter in accordance with at least some of the multiple email messages of the plurality.

14. The non-transitory computer-readable medium of claim 8 wherein providing the one or more indications of the aggregate sentiment information generated regarding the subset of multiple email messages is performed periodically in accordance with configuration information associated with at least some of the one or more recipient users.

15. A system, comprising:

one or more processors;

a hardware-implemented sentiment analysis engine;

a hardware-implemented sentiment aggregation engine, wherein the hardware-implemented message sentiment classification and prioritization system comprises a microprocessor of a special-purpose computer; and

at least one memory, the memory including instructions that, upon execution by at least one of the one or more processors, cause the system to perform a method for providing aggregate sentiment information regarding user email messages, the method comprising;

receiving one or more indications of a plurality of email messages sent to one or more recipient users;

analyzing, based at least in part on one or more natural-language processing libraries, multiple email messages of the plurality of email messages, wherein analyzing the multiple email messages includes;

determining one or more sentiments associated with each of the multiple email messages;

identifying one or more subsets of the multiple email messages, wherein each subset comprises a plurality of email messages and the email messages of each subset are complete email messages, the one or more subsets of the multiple email messages are based on a time period in which email messages were received and/or sent; and

generating, based at least in part on the determining of the one or more sentiments, aggregate sentiment information regarding the one or more subsets of the multiple email messages, wherein the aggregate sentiment information comprises an aggregation of sentiment information from the plurality of email messages of each subset; and

providing, for at least one subset of multiple email messages, one or more indications of the aggregate sentiment information generated regarding the at least one subset; and

displaying the one or more indications of the aggregate sentiment information via a display.

16. The system of claim 15 wherein the method is performed at least in part via a software application plug-in executed in conjunction with an email client associated with each of at least some of the one or more recipient users.

17. The system of claim 15 wherein generating the aggregate sentiment information regarding the subset of multiple email messages includes extracting one or more keywords associated with a sentiment that is determined to be associated with each of the email messages of the subset.

18. The system of claim 15 wherein providing the one or more indications of the aggregate sentiment information generated regarding the subset of multiple email messages includes displaying statistical data regarding one or more sentiments associated with the multiple email messages of the plurality.

19. The system of claim 15 wherein providing the one or more indications of the aggregate sentiment information generated regarding the subset of multiple email messages includes providing at least some of the aggregate sentiment information to one or more users other than the one or more recipient users.

20. The system of claim 15 wherein the method further comprises, subsequent to the generating of the aggregate sentiment information, receiving a search parameter and responding by providing an indication of one or more sentiments determined to be associated with the search parameter in accordance with at least some of the multiple email messages of the plurality.